CENT PAPER NUMBER ONE (A):

RETHINKING THE PSYCHOLOGICAL MODELS UNDERPINNING

RATIONAL EMOTIVE BEHAVIOUR THERAPY (REBT)

Copyright (c) Dr Jim Byrne, December 2009

1. Introductory Comments

Cognitive Emotive Narrative Therapy (CENT) arose out of my attempts to

reconcile Rational Emotive Behaviour Therapy (REBT) and certain other
elements of therapy systems that I found useful: commencing with
Transactional Analysis (TA), and Zen philosophy. It was also shaped by my
discovery of some limitations of certain aspects of REBT theory. However,
much of the foundations of REBT still serve as important elements of
CENT.

Over the period 1999-2005 I was in correspondence with Dr Albert Ellis,

the creator of REBT, concerning my thinking about various aspects of
REBT theory and practice, and I was always totally open about those
aspects that I found most helpful, and those aspects that caused me some
concern. For examples: I was not convinced that 'Socratic Questioning'
was always a helpful process; I felt that REBT did not have an adequate
'theory of personality'; I had some ideas about how to scientifically
investigate the effectiveness of REBT; and so on.

Before I tackle the substance of this paper, which is a defence of the core
model of REBT from some unsupportable claims made by Bond and
Dryden (1996), I would like to review my take on the strengths of Rational
Emotive Behaviour Therapy.

The great strength of Rational Emotive Behaviour Therapy (REBT) is that it

teaches a philosophy of life. In a nutshell, this is it:

"If you will just give up ‘awfulizing', ‘demandingness', ‘low frustration

tolerance', and ‘condemning and damning yourself, others and the world',
then your life will become calmer and happier".
Let us now look at what is involved in implementing such a philosophy in
your own life.

To give up awfulizing you have to develop a way of accurately rating the

degree of badness of those noxious events and situations by which you are
beset. The best system for this kind of accurate rating, that I have seen, is
that developed by Dr Tom Miller. In Self Discipline and Emotional Control
(which is available online as an audio or video programme), he developed
something called The Johnny Carson Scale which helps us to put a realistic,
numerical degree of badness on any adversity with which we are faced. By
becoming more realistic, be also become less disturbed.

To give up demandingness is a more complex problem. In moderate

forms of REBT, giving up demandingness means giving up extreme forms
of demands, such as: "I must be loved and approved by all significant
others, all of the time, and if I'm not, then this is awful, and I am no
good". This is clearly illogical, and unreasonable, and for this reason we
say it is irrational. In extreme forms of REBT, on the other hand, all forms
of usage of the words "should", "must", "have to", "ought to", "got to",
"need to" are outlawed, proscribed and prohibited to ourselves - except
for their use to describe "reality". For example, I would be able to say:
"My life should be the way it is, because it is!" Unfortunately, this means
we cannot make any moral prescriptions, or realistic demands upon
ourselves, others and the world. (Although Tom Miller is very good at
teaching how to get rid of awfulizing, he is an extremist when it comes to
eliminating demandingness!) My own view is this: We should give up all
unrealistic and unreasonable demands on ourselves, others and the world;
but keep our moral prescriptions and our reasonable demands about our
own behaviour, the behaviour of others, and the nature of the social
world. (To learn my form of giving up demandingness, see in particular my
pamphlet: Overcoming Fear and Anxiety).

To give up low frustration tolerance we need to develop persistence, a

commitment to work steadily towards our goals, and a system of rewards
and penalties to keep our show on the road. And we need to maintain our
lives in reasonable balance. In particular, we need to focus on the costs of
quitting under pressure, and the benefits of soldiering on despite all
setbacks, frustrations and difficulties. (See my Overcoming
Procrastination programme).
To give up condemning and damning of ourselves, others and the world,
we need to develop our sense that "I'm okay, and you're okay" - in the
sense that you and I are totally acceptable to me exactly as we are, even
though we might want to change our own or each others behaviours. And
we need to learn to manage our emotional reactions to the world. (See my
Supreme Self-Confidence book). However, in extreme forms of REBT, this
philosophy is then expanded to include "unconditional self acceptance",
which mirrors Carl Rogers "unconditional positive regard". However, as I
have argued elsewhere, as long as we are social beings, living in social
communities, be cannot extent "unconditional acceptance" to anybody, as
their internalized systems of self-regulation depend upon appropriate
guilt when they contemplate immoral acts; and appropriate shame when
they consider how "we" will judge them when they engage in those
acts. Because we accept them conditionally, so long as they act morally,
they are likely to stive to act morally. As soon as we accept them "no
matter what they do, even planning and carrying out a Holacaust", many of
them will feel freee to unleash their "bad wolf" on all of us. So
"unconditional self and other acceptance" is completely unacceptable to
me, and in Cognitive Emotive Narrative Therapy (CENT), I have replaced it
with "one-conditional self and other acceptance", which means, I will
accept you, and myself, as long as we strive to always act morally; and as
long as we resist any and all temptations to act immorally.

~~~

This paper now takes up some elements of an argument mounted against

certain aspects of REBT theory in 1996 by Windy Dryden and Frank Bond -
Bond and Dryden (1996) - and shows that Albert Ellis had a much more
complex model of mind than they had appreciated; and that I had
identified some critical errors in the thinking of Bond and Dryden.

The central model used in REBT is the ABC model, which is normally
presented like this:

A = An Activating event, or stimulus, which results in some kind of

response from an individual.

B = The Belief system of the individual (which includes distinctions to do

with whether the individual: (a) is being ‘demanding' or merely ‘preferring'
something; (b) is ‘awfulizing' or merely saying something is some small
degree of badness; (c) is implying that they cannot stand something or
merely that it is difficult to stand it; or (d) they are condemning self,
others or the world, or merely being critical of their own behaviour, the
behaviour of others, and/or some features of the world/reality).

C = The Consequent emotions and/or behaviours that arise out of the

interaction of the ‘A' times the ‘B' above. The implication is that an
extreme belief times a difficult stimulus will result in an intense emotional
response; while a moderate belief times the same difficult stimulus will
result in a greatly reduced emotional response. Hence, therapy consists of
helping the client to develop moderate beliefs.

(The core of the remainder of this article is abstracted, with minor

amendments, from a paper I wrote in 2003 - which is listed in the
References below as Byrne (2003b).)

2. Bond and Dryden's Major Error

The first unhelpful aspect of Bond and Dryden (1996) was that they
conflated the content of the arguments in Ellis (1958) - concerning the
ways in which cognition and emotion "are not two entirely different
processes, but that they significantly overlap in many respects " - and the
Ellis (1994) arguments concerning the complexities of the As, Bs, Cs and
their interactions. These discussions concern quite distinct aspects of
human functioning, in my reading and interpretation. The first (or 1958
presentation) is concerned with how thinking and feeling interact in the
processing of incoming stimuli - which in the later-emerging A>B>C model
would place them as overlapping processes at point B. The second (or
1994 presentation) extends the discussion of overlapping phenomena to
include how the A's, B's and C's each overlap and interact with each
other. This interpretation seems to me to be supported by the following
statement from Ellis (1994), page 80:

"Perhaps the main thing I want to emphasize in this (1994) chapter is that
not only, as I have previously theorized (in 1958-62), are cognitions,
emotions, and behaviours interactional, and not only are they practically
never entirely disparate and pure, but the same thing seems to go for the
ABC's of REBT." (Dates in brackets added for clarification - JWB).

What this means is that, in Ellis (1958) - and incidentally also in Ellis 1962 -
Albert Ellis is describing how thinking and feeling are overlapping and
interactional, at the level of stimulus processing, but he has not yet
posited the A>B>C model. If the ABC model had existed in 1958, I am
confident that Albert Ellis would have said "thinking and feeling are
overlapping processes at point B in the model". However, in Ellis (1994),
he has moved on to look at the ways in which (now that the A>B>C model
has existed for more than two decades) all three elements of the A>B>C
model influence each other. Bond and Dryden do not acknowledge this
distinction, indicating that they have misunderstood the 1958
argument. Instead, they "invent" something which they call an
"interdependency principle" which they say Ellis has proposed as his
explanation of the interactions between the As, Bs and Cs; and they
overlook the very strong probability that what Ellis (1958) is referring to is
interaction of thinking and emotion within point B of the later model,
which did not exist at that time. (In Ellis [1958], [1962], [1994], and again,
later, in Ellis and Dryden [1997/1999] the concept which is used is
"interaction" and "interactional", not "interdependent" or
"interdependency". [See in particular, Ellis and Dryden, 1997/1999, page 5,
second and third paragraphs]). So this label, created by Bond and Dryden
(1996) - the so-called "interdependence principle" - is not a phrase used by
Albert Ellis, in 1958 or 1962; nor in 1994. Furthermore, this label is
confusing and misleading, if only for the reason that I have been able to
distinguish four types of interdependency, ranging from a weak form to an
ultra-strong form. (Byrne, 2003, page 7). So I want to start by going back
to the 1958 presentation - Ellis, 1958 - of the ways in which thinking and
feeling are said by Albert Ellis to be interrelated.

"Rational psychotherapy is based on the assumption that thought and

emotion are not two entirely different processes, but that they
significantly overlap in many respects and that therefore disordered
emotions can often (though not always) be ameliorated by changing one's
thinking".

...

"It is also hypothesized that among adult humans raised in a social culture,
thinking and emoting are so closely interrelated that they usually
accompany each other, act in a cause-and-effect relationship, and in
certain (although hardly all) respects are essentially the same thing, so
that one's thinking becomes one's emotion and emoting becomes one's
thought". (Page 36).
If we want to model this proposition, in my view, then this is how we
would do it. Firstly, we would review the S>O>R model of the neo-
behaviourists as follows, as this model did exist at that time. (The neo-
behaviourists, of course, were part of the "cognitive turn", or "cognitive
revolution" against the stimulus>response model of the classical and
operant behaviourists).

Figure 1: The Stimulus>Organism>Response Model

In the S>O>R model, most of the "reality" is contained in part ‘O': the
organism itself. The ‘S' is frequently seen as an "incoming stimulus",
though in humans in particular, the "internally generated stimuli" are just
as important. And also, the ‘S' is largely an interpretative inference created
in the mind of the human subject. So, again in the case of humans, even
part of the ‘S' is essentially contained within the ‘O'. The ‘R' is
conceptualized as the response of the organism, such as a rat jumping
upon sensing an electrical current in the floor of its cage. In the case of
humans, the ‘R' is partly observable behaviour of a gross physical kind,
such as running away, punching, kicking; partly more subtle physical
behaviour, such as withdrawing from society; and partly even more subtle
non-verbal behaviours, such as squinting, smiling, scowling, teeth-baring,
and so on. But underlying those visible "outputs", at point ‘R', are a whole
load of invisible electro-chemical outputs, and cardio-vascular outputs,
and hormonal outputs, and so on.
So again, much of the ‘R' is invisible, and contained "inside" of the
organism (‘O'); while the visible "outputs" are technically "outside" of the
organism, but still non-separate manifestations of the activation of the
organism. So even with the S and the R, we can never find a 'response'
"separated" from the organism. Now please note that none of this
"interactional nature" of the S>O>R model prevented the neo-
behaviourists from designing and developing ingenious experiments to
test their theories of how organisms were "wired up" at ‘O'. And why
not? Because: while the S, O and R are not separate from each other, they
are distinct phases in a process of sensing, processing and outputting. They
can be distinguished from each other. Relations between them can be
inferred and hypothesized. Predictions can be made - on the basis of
theories about what goes on inside ‘O' - about what will happen if the
stimulus (‘S') is changed in some kind of way.

Experiments can be designed and implemented to see if these predictions

come to pass. If they come to pass as predicted, then the hypothesis
about the relation of S to O, and the internal processes of ‘O', can be said
to have been indirectly validated! They also have not been
invalidated. This process of prediction and testing; of producing indirect
validation and demonstrating that invalidation has not occurred, is as good
as science gets in this kind of investigation.

3. From S>O>R to A>B>C

The S>O>R model preceded the A>B>C model, and there is clearly a close
fit between the three elements of the two models. The stimulus (S) is
equivalent to the Activating Event (A), while the response (R) is equivalent
to the Consequence (C). What I want to emphasize here is that in the
S>O>R model, the O is the whole organism, human or otherwise. In the
A>B>C model, by association, we can think of the B as the whole organism,
with the emphasis on the belief system. And again by extension we can
note that Ellis (1958) is proposing that the organism has "overlapping"
thinking and feeling processes (as well as behavioural processing
capabilities). Now, even though he does not present a model, in 1958, and
does not say "where" that overlapping takes place, my interpretation is
that it takes place at point B in the later-developed A>B>C model, as
follows:
 Figure 2: The overlapping of thinking and feeling in Ellis (1958)

The reason I have included "attitudes" in point B of the model in Figure 2,

above, is that, Ellis (1976) hypothesizes that "irrational beliefs" are
inborn. Since "beliefs" are normally assumed to be language-based
propositions, and language-based propositions cannot, by definition, be
inborn, I infer that the inborn form of the irrational beliefs is that of
"attitudes", or "attitudinal orientations", including "demandingness" and
"awfulizing".

If this model is correct, and I think it probably is, then does this in any way
prevent us inferring that irrational beliefs, at point B in the model,
determine or cause unhealthy negative emotions at point C, as suggested
by Bond and Dryden, 1996? Not at all; if we recognize that "irrational
beliefs" is shorthand for "irrational cognitive/emotive/behavioural
attitudes and beliefs". The fact that thinking and feeling interact at the
level of "processing" within the human organism, and seem to be in many
respects "essentially the same kind of phenomenon", does not in any way
compromise the model which states that this cognitive-emotive processing
(at point B) is responsible for the resulting emotional-behavioural outputs
at point C in the model. When that cognitive-emotive processing is
rational, we get reasonable upsets (emotional-behavioural) at point
C. When that cognitive-emotional processing at point B is irrational, we
get overly-upset emotions-behaviours at point C.

(Much of the confusion in Bond and Dryden [1996] may be caused by the
fact that they have not realized that, although adult thinking-feeling
processing, at point B, comes to be dominated by language from about the
age of seven years, and especially from puberty onwards - as argued by
Vygotsky - this does not remove or eliminate the underlying electro-
chemical processing of synaptic signals in the cognitive and emotive
centres of the brain. It also does not remove or eliminate the non-verbal,
visual schemas which underlie the verbal; nor the kinaesthetic schemas
which underlie the visual - as argued by Bruner. [See Bruner and Vygotsky,
in Wood, 1988/1994]).

4. REBT and Cognitive Psychology

The next problem for me is how to relate this model to existing

psychological models within the mainstream. Elsewhere, I have developed
a model which I will present shortly. But first, I want to deal with a
proposal from David (2003) to place REBT "...more explicitly in the context
of contemporary cognitive psychology". I agree with this proposal in
principle, as cognitive psychology has as its foundation a process of
inferring psychological processes and structures from the observable
behaviour of research participants, and then testing those inferences in
practical experiments. (Eysenck and Keane, 2001). This, it seems to me, is
preferable to the flawed approach of trying to extract statements of
"belief" from research participants, for the very good reason that research
participants do not always, or even normally know what internal process
or processes (at point B) gave rise to their emotions and behaviours at
point C in the A>B>C model. ("Nisbett and Wilson [1977, p.248] claimed
that people are generally unaware of the processes influencing their
behaviour..." - Eysenck and Keane, 2001, page 4). However, there is a
practical problem with David's (2003) proposal, in relation to my present
task. Cognitive psychology has a long history of largely ignoring the
significance of emotion, since it had its origin in information processing
theory, based upon the functioning of post-war computers. (Eysenck and
Keane, 2001). Over its first fifty years, cognitive psychology has
concentrated mainly upon attention, perception, memory, language and
learning. In recent years there have been growing attempts to expand the
coverage of emotional issues addressed in cognitive psychology, but that
process is in its infancy. (Less than 5% of Eysenck and Keane, 2001, is
devoted to the connection between thinking and feeling, and there are no
explicit, general models of those connections presented). Two of the
most notable contributors to this growing edge of cognitive-emotive
enquiry in cognitive psychology are Antonio Damasio and Joseph LeDoux.
(Cf: Damasio, 1994 and 2000; and LeDoux, 1996). However, other
contributors include Zajonc (1980, 1984) and Lazarus (1982).

For the reasons outlined above, I propose to develop an updated A>B>C

model which is grounded upon the sustainable aspects of Freud's analysis
of the components of human personality, augmented by the best of
developmental and cognitive psychology. (This does not mean that I
support psychodynamic approaches to psychotherapy practice, but rather
that I consider that the "divisions of the mind" established by Freud still
have considerable validity, at least in their structural aspects).

Even if we assume that Ellis 1958 is sometimes referring to cognitive

processing, on the one hand, and emotive output, on the other, and how
the first determines the second, we could not sustain the proposition that
he was saying that processing was always cognitive. And again, although
he had not begun to present his ideas in the A>B>C format, I think the
following model takes account of most of his 1958-1962 presentations
(with the exception of the role of behaviour/physiological-arousal at the
processing level):
 Figure 3: The complex interaction of thinking and feeling, at point B;

and the interaction of the B and the C at point C1.

The philosophical/psychological justification for the distinction between

the A1 and the A2 is fully explored in Byrne (2003a). The overlapping of
the B and the C will be fully explored later in this paper.

This is how I conceptualize Ellis's statement about emotions and thinking

being overlapping and (practically/essentially) the same thing:
 Figure 4: Schematic illustration of the "significantly overlapping"
cognitive and emotive processes, both of which exist within point B of
the A>B>C model.

The B1 in Figure 4 is Ellis's (1976) inborn rational/irrational belief system,

plus other inborn attitudes which are necessary to account for human
personality. The B2 is all those cultural forms which are modelled for us,
and taught to us by our moral instructors (parents, siblings, teachers,
peers, neighbours, police, and so on). While the B3 is our self-constructed
(ego) personality.

Taken together, these overlapping cognitive-emotive processing

capabilities constitute the core of the B - or belief system - of the
organism. So where do the "other two basic life processes" fit in:
 Figure 5: The A>B>C model related to time and
sensing>processing>moving

"Sensing" fits in at point A. "Moving" fits in at point C, and it could be

argued that the evolutionarily valuable function of emotion is that it
provides the urge towards those forms of locomotion that have allowed
certain species to survive. (In the case of humans, much "moving", at point
C, has become symbolic: smiling; nodding; shaking hands, and so
on). Those species which arose, but did not have highly reactive
sensing>processing>moving "equipment" or "software" built into their
central nervous systems may well have been eaten by their more alert
predators. Organisms which responded with curiosity, or "love", to the
arrival of a potential predator would not have been "well adapted" to
survival. This insight suggests the following amendment to the A>B>C
model:
 Figure 6: Behavioural repertoires are partly inherited,

partly culturally transmitted and partly self-constructed at B3.

Figure 6 indicates that Ellis (1958) and (1962) considers that

behaviour/arousal is just as important a part of the processing of incoming
signals as are emotion and cognition. But the presentation in Figure 6,
while formally correct, is static and unhelpful in understanding these
relationships.

When I first started to teach REBT to others, in 1992-93, and later when I
began to work in private practice as an REBT therapist, in 1999, I was
constantly striving to find a way to visually represent human emotion in a
way which would make it understandable to my clients, and to allow them
to intervene in their own processes. In addition to presenting various
forms of the A>B>C model, I also developed something called "the Y-
model", to illustrate the kinaesthetic/bodily aspect of emotion
arousal. Initially this was rather crude, but after receiving developmental
feedback from my supervisor, I eventually produced Figure 7 below:
Figure 7: Jim Byrne's Y-Model of Emotion linked to Physiological Arousal

The Y-model shows all the major emotions dealt with in REBT therapy,
plotted on a vertical scale of physiological arousal - with extreme forms of
depression, hurt, guilt and shame at the minimum level of physiological
arousal, and the most extreme forms of anxiety and anger at the maximum
level of physiological arousal. Each individual is assumed to have their
own baseline, or resting level of arousal. (Gottman [1997], page 25). I do
not propose to go through this model in detail here, merely to state that a
depressed individual behaves very differently, kinaesthetically/
behaviourally, than an angry person. And a concerned individual, for
example, does not get as aroused as a person who is feeling anxious. Thus
visceral/facial/kinaesthetic/bodily arousal is part of the process of
producing and experiencing emotion arousal. Furthermore, each of the
states of emotion arousal in Figure 7 is assumed to be caused by the
interaction of (1) a particular A2 (perception/ interpretation/ inference);
plus (2) a particular overlapping cognitive-emotive "belief/attitude",
classified as rB (rational belief) or iB (irrational belief). Let us now revisit
the A>B>C model, as combined with the Y-model.
 Figure 8: The A>B>C Model Related to the Y-Model

Figure 8 shows a weight lifter, thinking-feeling-behaving in relation to his

task. This image suggests that, when something happens at A1, it is
interpreted at A2 (not shown), which triggers cognitive-emotive
processing of the A2 signal at B (1, 2 and 3). At the same time, the B1
(unconscious cognitive-emotive processing) sends a signal to the Y-model
(visceral, facial, physiological arousal), which responds by sending a signal
to the C1 (not shown) where it combines with the output from B, and
together these signals produce the emotional-behavioural response at
C. As it stands this could seem to be a fairly straight restatement of the
James-Lange theory of emotion. (Kagan and Segal, 1992, pages 321-322).

Of course, it could be objected that Walter Cannon put a major fly in my

ointment with his experiments which seemed to show that an
experimental animal surgically separated form its viscera, could still feel
fear. However, Cannon and Bard's theory does not cause me much
concern. They considered "...the physiological changes to be a sort of side
effect - useful in preparing the body to take appropriate action but not
essential to our conscious experience of emotion". (Kagan and Segal, 1992,
page 324). My own view is closer to that of Nicolas Humphrey (1992), who
got around the problem of Cannon and Bard's reservations by working on
the mystery of phantom limb syndrome. An individual who has lost his
hand may occasionally feel strong sensations in the "missing hand", and
want to scratch it. This phenomenon goes on for many years after the
hand was lost, and possibly for the remainder of the individual's
life. Humphrey (1992) resolved this conundrum by realizing that, through
a process of evolution, an original system of sending a signal from body
surface to brain, and a response back to the original part of the body
surface, the brain had learned to create brain-based representations of
the source of the stimulus, and to maintain two loops. The first, a long
one, runs from the body surface to the brain and back again, while the
second, a short one, runs from the sensing part of the brain, to the brain-
based representation of that part of the body. Thus a person may have a
missing hand, but they still have a brain-based representation of that hand,
which can "feel" sensations. By corollary, it is my contention that the
brain contains stored representations of what it was like - (at Y - some
level of physiological arousal) - to be anxiously aroused, angrily aroused
and so on. When the A1 triggers the A2, the A2 triggers the B1/2/3, the
B1/3 sends two signals (a la Cannon-Bard) - one to the viscera to signal the
need for physiological arousal, one to the C1 to initiate emotional affect -
but unlike Cannon-Bard, and very much in line with Humphrey (1992) a
third signal is sent to a "mental Y", or brain-based representation of what
it was like to have the physiological arousal (Y) which is "appropriate" to
this particular A2/B1-3 (cognitive-emotive arousal). Thus James-Lange
and Cannon-Bard are reconciled by "Humphrey-Byrne" with the addition of
a third signal which goes to a brain-based representation of some
historically stored physiological arousal. This can be shown visually as in
Figure 9, below:
 Figure 9 - The A>B>Y>y>C Model al la Humphrey-Byrne Linkages

Using the model in Figure 9, we would say that a stimulus at A1 triggers an

interpretive inference at A2, which triggers an evaluative (cognitive-
emotive) processing of that signal at B1-3. The B1 or 2 or 3 (to be
discussed later) then sends out at least three signals. One goes to the
face-body to initiate a physiological response. One goes to the collection
of little internal Y's, (inside the B boundary) representing physiological
responses to previous stimuli of the type apprehended at A2. And one
goes straight to the C1 to signal an affective output. The brain-based
little y's yield up the best fit to A2 of historically appropriate physiological
arousal, and send a signal to C1 containing that information. (This is similar
to what is called 'pattern matching' in the 'Human Givens' tradition: Griffin
and Tyrrell, 2004). It is anticipated that the signal from B1/2/3 and the
signal from the little internal Y's will meet and combine at C1, and initiate
the affective output at C2. Perhaps one second later, the feedback signal
from the bodily Y will reach C1, and reinforce the output at C2. And the
hormonal responses, running behind the electrical signals, will further
reinforce that message, perhaps another second later. All of these
processes are governed by the fundamental principle that the organism,
including the human organism, is a creature of habit, and outputs are
selected on the basis of best fit from a large collection of historically
accumulated representations in long term memory. Habitual ways of
responding may be broken after reflection upon them, but not in the
process of responding to a new stimulus. (This was the only truth in the
old behaviourist theory of Stimulus>Response>No Choice! Free will in
humans is mainly "Free won't - after the event!" We can dispute our iB's,
once we find out what they are, but we cannot eliminate them in advance!)

What has been achieved in Figure 9 is a visual schematic representation of

Albert Ellis's statement, cited above, from Ellis (1994):

"Perhaps the main thing I want to emphasize in this (1994) chapter is that
not only, as I have previously theorized (in 1958-62), are cognitions,
emotions, and behaviours interactional, and not only are they practically
never entirely disparate and pure, but the same thing seems to go for the
ABC's of REBT." (Dates in brackets added for clarification - JWB).

What the preceding six schematics, in Figures 4 to 9, illustrate most clearly

is this: It is very easy to misunderstand any discussion which does not have
physical or visual referents. Communication is difficult. This is so because
language is so "slippery". Schematic diagrams can help to clarify issues, if
only by giving you referents to which you can point when tracking the
elements of any particular system. (Korzybski [1933] is a good example of
this point, where he introduces his visual heuristic, called the "differential
analyzer"). And in my schematics above, we see that it is important to
distinguish between "emotional processing" (at B) and "emotion as an
output" (at C). (The total C is Emotional and Behavioural selection and
outputting). Emotional and cognitive processing, at B, are essentially
overlapping, and in some respects remarkably similar types of phenomena
- and they include behavioural/kinaesthetic/physiological components,
(YYYYY), in the form of brain-based representations of historically
accumulated experiences of emotion-physiology arousal. But emotional
output at C is quite distinct from emotional processing at point B in the
model. Furthermore, it is obvious when we incorporate a time-line into
the A>B>C model that C (at time t3) cannot under any circumstances
impact, influence, or determine A (at time t1) or B (at time t2). What it can
influence is subsequent A's, B's and C's! And again, behaviour cannot be
outputted at point C unless there are already behavioural repertoires at
point B, which significantly overlap the cognitive-emotive belief system,
and provide a basis for habitual pairings of stimulus-processing-response,
in which the response is not just a selected emotion, but a linked
behaviour.

Ellis (1958) continues:

"It is finally hypothesized that since man is a uniquely sign-, symbol-, and
language-creating animal, both thinking, and emoting tend to take the
form of self-talk or internalized sentences; and that, for all practical
purposes, the sentences that human beings keep telling themselves are or
become their thoughts and emotions". (Page 36).

This is a surprisingly sound statement for 1958, since Piaget had

misunderstood the significance of language, and the first translation of
Vygotsky's refutation of Piaget's errors was not published in the United
States until 1962, with an introduction by Bruner. Vygotsky argued, and
was supported by Bruner in this view, that thinking and language arise
separately, and only very slowly combine in the growing infant. (Cf: Wood,
1988/94). Eventually, however, by about the age of seven years, language
comes to dominate the thinking of each individual. But even after
language comes to constitute conscious thinking in each individual, it is
important to remember that, underlying this linguistic thinking, there are
visual and kinaesthetic forms of information processing. This insight
arises from Bruner's hypothesis that the earliest schemas which are laid
down by each infant are kinaesthetic. Next come visual schemas. And
finally language-based schemas. So, post Vygotsky/Bruner, it is safer to
say that while emotional and cognitive processing can be expressed in
language, the processing per se is electro-chemical. And in between the
electro-chemical level and the linguistic translation, there may be layers of
kinaesthetic and visual schemas which can also be, and normally are,
translated "upwards" into linguistic statements; just as, by the same token,
linguistic statements can be translated downwards, via various types of
schemas, into various types of cognitive-emotive-behavioural arousal at
the electro-chemical level. It is in this sense that it is valid to ask: "What
are you telling yourself to cause yourself to feel what you feel?" You are
either telling yourself something, linguistically, at conscious or
preconscious levels; or you are signalling yourself equivalent messages
visually, kinaesthetically (at pre-conscious or unconscious levels), or
electro-chemically (at unconscious levels). Ellis (1958) says nothing that
contradicts this more updated view, and he says much that foreshadows
these later developments.

Ellis (1958) goes on to admit that emotional outputs can exist without
thought - but only momentarily. This is another very good example of Ellis
(1958) foreshadowing much later developments. This statement fits in
with LeDoux's (1996) view of the unconscious causation of automatic
emotions (at C2) triggered via the B1. (However, this idea was already
present in Arnold [1960] who was thoroughly reviewed by Ellis
[1962]). But what Ellis is saying is that, the maintenance of such an
emotion depends upon subsequent appraisals of the output at C2, (at a
subsequent A2), which is precisely what modern ‘cognitive appraisal
theory' contends. Ellis's statement anticipates the work of Schachter and
Singer [1962] by four years. And indeed, Schachter and Singer, like so
many after them, may simply have been ploughing Ellis's 1958 and 1962
furrows even deeper.

Finally, to sum up: I think Ellis (1958) was saying that an incoming stimulus,
being encountered by a human organism, triggers cognitive-emotive-
kinaesthetic processing, which is essentially sign-, symbol-, and language
based processing, which results in an emotional-behavioural output which
is coloured or shaped by the nature (rational or irrational) of the cognitive-
emotive-behavioural processing. We can fit this into the later developed
A>B>C model as follows:

Figure 10: How to understand Ellis (1958)

In the model in Figure 10, the A is an external or internal stimulus calling

for a response from the human organism. This occurs at Time 1. One unit
of time later, the stimulus has been perceived, interpreted and
inferentially located in terms of the organism's experience (at A2). Then
at point B3, where the cognitive and emotive processing capabilities
overlap or interact, there is a quick evaluation of the stimulus, (at time t3)
which produces a habitual emotional-behavioural response at point C. The
fact that the cognitive and emotive processing capabilities at point B
overlap, or interact, does not in any way prevent the organism selecting an
appropriate response and outputting it at point C. Indeed, Damasio 1994 -
in describing the case of Elliot, who was incapable of making decisions,
due to damage to the emotional centres of his brain - has shown that an
emotional processing capability is necessary in order for a human to be
able to make a "cognitive" decision.

Furthermore, at point B3, there may have been a sub-vocal thought,

and/or a sub-conscious attitude, and/or an unconscious attitude, and/or an
unconscious electro-chemical selection process, implicated in the
cognitive-emotive processing that took place. I would hypothesize that
when we have plenty of time to think about an Activating event (A), we
are more likely to function consciously and linguistically. And when we are
under pressure to respond urgently, we tend to operate at the opposite
end of our functional scale, perhaps even totally unconsciously and non-
linguistically. (LeDoux, 1996; Korzybski, 1933; Ellis, 1958, 1962, 1994;
Damasio, 1994, 2000).

If this model can be supported, and if it had been accepted by Albert Ellis
as being a true reflection of his 1958-62 thinking, then there would be no
basis to Bond and Dryden's (1996) claim that, because of the way Ellis
(1958) construes the interactions of thinking, feeling and behaving, it
"...cannot be established that cognitions are at the core of psychological
disturbance and health (the core hypothesis)". (Bond and Dryden, 1996,
page 29). Indeed, if this model had been accepted by Albert Ellis as
representing his 1958-62 thinking, then there is no basis to Bond and
Dryden's claim that "the core hypothesis" of REBT is that "cognitions are
at the core of psychological disturbance and health". It is certainly my
view that the core hypothesis of REBT is that "Rational beliefs" and
"Irrational beliefs" - (which includes "rational" and "irrational" attitudes as
inborn orientations [Ellis, 1976]) - are responsible for how we appraise
incoming signals, and the emotional-behavioural outputs that we produce
in response to those signals. And while language dominates our cognitive-
emotive processing, it does not entirely or exclusively constitute
it. Indeed, very much more of our processing at point B in the A>B>C
model is non-verbal than that which is verbal. And although Albert Ellis
tends to emphasize what clients are "telling themselves" to make
themselves upset, his writings reveal a much richer description of
cognitive-emotive-behavioural functioning than is suggested in this
therapeutic shorthand.

Concluding Comment

The core of this paper was written in 2003 and never published, perhaps to
some extent because of the politics of REBT factional differences at that
time; but also because I wanted to retain control over the content of my
writings, and not have them emasculated by individuals who were out of
sympathy with my integrative approach. It is now being published by the
Institute for CENT studies as the foundation of that element of CENT
which derives from REBT. it is also intended as a posthumous acclaim of
Dr Ellis's grasp of the complexity of human functioning, which was often
allowed to drop out of the cut and thrust of REBT therapy sessions, and
out of simplistic, popular representations of this model of therapy. REBT
is a highly complex understanding of the highly complex human mind -
which is probably the most complex entity in the universe. CENT is
similarly complex, as a measure of the clients with whom we deal.

~~~

To reference this paper in a publication, please use the following citation:

Byrne, J. (2009h) Rethinking the psychological models underpinning

Rational Emotive Behaviour Therapy (REBT). Cent Paper No.1(a). Hebden
Bridge: The Institute for CENT Studies.

~~~

REFERENCES

Bond, F.W. and Dryden, W. (1996). Why Two, Central REBT Hypotheses
Appear Untestable. Journal of Rational-Emotive & Cognitive-Behavior
Therapy, 14(1), 29-40.

Byrne, J.W. (2003a). On the conceptual errors of Bond and Dryden, 1996:
or how to scientifically validate the central hypotheses of REBT,
Occasional Paper No.7, 94pp, Hebden Bridge: ABC Coaching Publications.

Byrne, J.W. (2003b). Further Validation of REBT: a therapist's

viewpoint. Hebden Bridge: ABC Coaching Publications.

Damasio, A. R. (1994). Descartes' Error: emotion, reason and the human

brain, London, Picador.

Damasio, A.R. (2000). The Feeling of What Happens: body, emotion and the
making of consciousness, London, Vintage.

David, D. (2003). Rational Emotive Behavior Therapy (REBT): the view of a

cognitive psychologist, in Dryden, W. (Ed). Rational Emotive Behaviour
Therapy: theoretical developments, Hove, East Sussex, Brunner-Routledge.
Page 131.
Ellis, A. (1958). Rational Psychotherapy, Journal of General Psychology, 59,
35-49.

Ellis A. (1962). Reason and Emotion in Psychotherapy, New York, Carol

Publishing.

Ellis, A. (1976). The biological basis of human irrationality. Journal of

Individual Psychology, 32, 145-168.

Ellis, A. (1994). Reason and Emotion in Psychotherapy: revised and updated,

New York, Carol Publishing Group.

Ellis, A. and Dryden, W. (1997/1999). The Practice of Rational Emotive

Behaviour Therapy, Second edition, London, Free Association Press.

Eysenck, M.W. and Keane, M.T. (2001). Cognitive Psychology: a student's

handbook, Hove, East Sussex, Psychology Press.

Gottman, John (1997). Why Marriages Succeed or Fail: and how you can
make yours last, London, Bloomsbury.

Griffin, J. and Tyrrell, I. (2004) Human Givens: A new approach to emotional

helath and clear thinking. Chalvington, East Sussex: HG Publishing.

Humphrey, Nicholas (1992). A History of the Mind, London, Vintage.

Kagan, J. and Segal, J. (1992). Psychology: an introduction, Seventh edition,

Forth Worth, Harcourt Brace Jovanovich College Publishers.

Korzybski, Alfred (1933/1990). Selections from Science and Sanity: an

introduction to non-aristotelian systems and general semantics, Englewood,
New Jersey, The International Non-Aristotelian Library Publishing
Company.

LeDoux, J. (1996). The Emotional Brain: the mysterious underpinnings of

emotional life, New York, Simon and Schuster.

Lazarus, R.S. (1982). Thoughts on the relations between emotion and

cognition. American Psychologist, 37, 1019-1024.

Vygotsky, L.S. (1962). Thought and Language. New York. Wiley.

Wood, David (1988/1994) How Children Think and Learn: the social contexts
of cognitive development, Oxford, Blackwell.

Zajonc, R.B. (1980). Feeling and thinking: preferences need no inferences.

American Psychologist, 35, 151-175.

Zajonc, R.B. (1984). On the primacy of affect. American Psychologist, 39,

117-123.

~~~

PS: There are seven papers on the subject of CENT therapy on the CENT
Institute page.

There is a video on the homepage entitled ‘What is CENT?'

And there is a new video on ‘Taking Responsibility' on the Life, Happiness

and Success Coaching page.

~~~

If you like this paper, then why not post it to your favourite social
networking site with this button:

~~~

If you would like to be kept up to date with developments at ABC

Coaching, then please sign up for the ABC Newsletter, which will be sent
out monthly from January 2010.

Just send a blank email with the word ‘Newsletter' in the subject line, to
jim.byrne@abc-counselling.com.

~~~

Finally, you might want to recommend this page to a friend or colleague,

using this button: